Deep Learning
Sparse Deep Learning for Time Series Data: Theory and Applications
Specifically, we lay the foundation within the Bayesian framework.
Decomposing Novel into Known: Part Concept Learning For 3D Novel Class Discovery
The key challenge of 3D NCD is that learned features by known class recognition are heavily biased and hinder generalization to novel classes.
The Shaped Transformer: Attention Models in the Infinite Depth-and-Width Limit
Finally, we show, through simulations, that the SDE provides a surprisingly good description of the corresponding finite-size model. We coin the name shaped Transformer for these architectural modifications.
MarioGPT: Open-Ended T ext2Level Generation through Large Language Models
Procedural Content Generation (PCG) is a technique to generate complex and diverse environments in an automated way.
Scattering Vision Transformer: Spectral Mixing Matters
Existing solutions often resort to down-sampling operations, such as pooling, to reduce computational cost. Unfortunately, such operations are non-invertible and can result in information loss.
